J. Chem. Eng. Data 1989, 34,
250
tional group, the presence of the CH, and COOC groups, and the interaction parameters given in ref 78, a mean estimation of the vapor concentrations is obtained of 0.019 (mean error of 5.1 % ) for methyl propanoate-ethanol and of 0.009 (mean error of 1.4% ) for methyl propanoate-propan-1-01, I n the aggregate, the prediction of the equilibrium data of these methyl propanoate-alkanol systems, both with the ASOG method and with the three cases chosen from the UNIFAC model, is good, with an overall mean error of less than 4% in every case. Neither method can therefore be viewed with more favor than the other, nor can any one case of the three presented by singled out in the case of the UNIFAC model since, with all certainty, the resutts of the predictions may differ greatly according to the family studied. The azeotrope found for the system xH5C,COOCH3 (1 x)C2H5(0H)corresponds to a value of x = y = 0.483and T = 345.58K, a concentration that differs by, approximately, 7 % from that presented by Horsley (5). The ASOG group contribution model predicts the azeotrope of the above-mentioned system under conditions of T = 345.70K and x = y = 0.422, with an error of 12.6%, while, in the three UNIFAC model cases studied, the azeotrope value closest to the experimental one is achieved, under conditions of T = 345.55 K and x = y = 0.505, with an error of 4.6% in the estimation of the said singular point.
+
Registry No. EtOH, 64-17-5: PrOH, 71-23-8; methyl propanoate. 55412-1.
250-252
Ortega, J.; Peira, J. A.; de Alfonson, C. Lat. Am. J. Chem. Eng. Appl. Chem. 1988, 16, 317. Ortega, J.; Ocbn, J.; Peiia, J. A.; de Alfonso, C.; Paz-Andrade, M. I.; Fernanadez, J. Can. J. Chem. Eng. 1987. 65, 982. Polak, J.; Lu, 8. C. Y. J. Chem. Eng. Data 1972, 77, 456. Horsley, L. H. Azeotroplc Date ; Advances in Chemistry Series 6; American Chemical Society: Washington, DC, 1952. TRC, Thermodynamic Research Center Data Project; Texas ABM University: College Station, TX, 1969. Rkidick, J. A.; Bunger, W. 6.; Sakano. T. K. Organic Solvents. Techniques of Chemistry, 4th ed.: Wley-Interscience: New York, 1986: VOl. 11. Wilhoit, R . C.; Zwolinski, 8. J. Physicaland Thermodynarnlc Propedles of Aliphatic Alcohols; J. Phys. Chem. Ref. Data, 1973, 2. Boublick, T.; Fried, V.; Hla, E. The Vapour Pressures of Pure Substances: Elsevier: Amsterdam, 1973. Ortega, J.; Matos, J. S.; Par-Andrade, M. I.; Jimenez, E. J. Chem. Thermodyn. 1985, 17, 1127. Prausnitz, J. M.: Lichtenthaler, R. N.; Gomes de Azevedo, E. Molecular Thermodynamics of NuM-phase Equlllbrla, 2nd ed.;Prentice-Hall: Englewood Cliffs, NJ, 1986. Plzer, K. S.;Curl, R. F., Jr. J. Am. Chem. Soc. 1957, 79, 2369. Tsonopoulos, C. AIChE J. 1974, 2 0 , 263; IbM. 1978, 2 4 , 1112. Tarakad, R. R.; Danner, R. P. AIChE J. 1977, 23(5). 685. Spencer, C. F.; Danner, R. P. J. Chem. Eng. Data 1972, 17, 236. Kister, A. T. Chem. Eng. frog., Symp. Ser. 1952, 48(2). Redllch, 0.; 49. Herlngton, E. F. G. J. Inst. Pet. 1951, 3 7 , 457. Fredenslund, Aa.; Gmehling, J.; Rasmussen, P. Vapour-UquM Equllibria Using UNIFAC. A Group-Contribution Methcd; Elsevler: Amsterdam, 1977. Ortega, J. J. Chem. Eng. Data 1985, 3 0 , 465. Kojima, K.; Tochigl, K. Prediction of Vapow-Llquhi Equilibria by the ASOG Method; Kodansha Ltd.: Tokyo, 1979. Fredensiund, Aa.; Jones, R. L.; Prausnitz, J. M. AIChE J. 1975, 21(6), 1086. KjoWorgensen, S.; Kolbe, 6.; Gmehling. J.; Rasmussen, P. Ind. Eng. Chem. Process Des. Dev. 1979, 18(4), 714.
Literature Cited (1) Ortega, J.: Pefia, J. A.; de Alfonso, C. J. Chem. Eng. Data 1988, 3 1 ,
339.
Received for review August 4, 1988. Accepted January 17, 1989. We gratefully acknowledge the financial support of the Canarias Government for Project No. 52/86.
Gas Solubilities (H,, He, N,, CO, O,, Ar, CO,) in Organic Liquids at 293.2 K Petra Luhring and Adrian Schumpe" Technische Chemie, Universitat Oldenburg, 0-2900Oldenburg, West Germany
Soiubiilties of hydrogen, helium, nitrogen, carbon monoxide, oxygen, and carbon dioxide in 25 pure organic liquids and In 2 binary mixtures have been determlned at the temperature of 293.2 K. The results are compared to the avaliabie literature data, to the regular-solution theory, and to the scaled-particle theory. Experimental gas solubilities in organic liquids are tabulated for many gashiquid systems ( 1 ) . However, in a recent study (2)on oxygen diffusivities in organic liquids, some of the data needed to evaluate the diffusivities from the measured transmissibilities were not available. The regular-solutiontheory and the scaled-particle theory were not always applicable, and in other cases their predictions differed considerably. Therefore, gas solubilities in organic liquids were studied experimentally in 89 different gaslliquid systems and compared to the two models. * Address correspondence to this author at Qesellschaft fijr Biotechnologische Forschung (GBF), Maschercder Weg 1, D-3300 Braunschweig, West Germany.
0021-9568/89/ 1734-0250$01.50/0
Experimental Section The gas solubilities were determined by a barometric method used already in a previous study (3).The measuring chamber was a glass vessel divided into chambers for the liquid (V, = 349.6 cm3)and the gas (V, = 598.9cm3) by a horizontal glass plate with openings at the center and near the wail. The plate enabled the exact adjustment of the liquid level and inhibited premature gas absorption. The vessel was kept at a temperature of 293.2 f 0.1 K by means of a jacket connected to a thermostat; furthermore, the apparatus was placed in a box with an internal air temperature of 293.2 f 0.2 K. The substances were obtained from Merck at the highest available purity except for ligroin (alkanes with a bp range of 373-413 K). A surplus volume of liquid was degassed by evacuation. The process was terminated once the correct liquid level was reached by evaporation of the liquid at its vapor pressure P Dry gas of atmospheric pressure Po was slowly introduced into the head space. After pressure and temperature equilibration, the gas llne was disconnected and a magnetic stirrer in the liquid was started. The liquid overflowed the plate and got
,.
@ 1989 American Chemical Society
Journal of Chemical and Engineering Data, Vol. 34,
Table I. Henry's Constants for H2, 02,and C 0 2 a t H,Pa m3 mol-' liquid H2 02 acetone 27 080 9 400 ani1ine 93 870 38 480 benzene 11080 33 900 11390 1-butanol 34 870 9 360 n-butyl acetate 28 430 8020 carbon tetrachloride 32 130 8 620 cyc1ohexane 29 690 12 700 decalin 42 343 15340 1,2-dichloroethane 44 900 16 320 1,4-dioxane 49 110 10360 ethanol 29 380 8 600 ethyl acetate 29 290 11160 ethylbenzene 37 480 ligroin 7 558 29 740 10490 methanol 21 970 72 370 nitrobenzene 9 780 30 910 2-propanol 10720 35 960 1-tetradecene tetralin 16850 51 480 12 140 39 020 1,2,4-trimethylbenzene toluene 10 670 35 960 m-xylene 10890 34 700
293.2
K
Table 111. Henry's Constants for Oxygen i n Mixtures of Organic Liquids at 293.2 K cyclohexane/ benzene mixtures ethanol/toluene mixtures mole fracn mole fracn cyclohexane H,Pa m3 mol-' ethanol H,Pa m3 mol-' 0 11080 0 10670 0.067 10550 0.050 10920 0.134 9 990 0.136 10950 10 880 0.315 9 410 0.280 10770 0.540 8 850 0.550 10720 0.785 8 430 0.735 10540 0.900 8 540 0.885 1 10380 8 620 0.943 10 360 1
COZ 347 1702 946 1179 499 909 1414 2001 650 357 802 399 1124
601 905 1070 1551 1610 1251 965 1067
Table 11. Henry's Constants for He, Ar, N2,and CO at 293.2
K H,Pa m3 mol-' liquid meth ano1 1,2-dichloroethane toluene
He 85500 156 300 118000
Ar 10330 13940 10190
N, 17310 27 640 19060
co 12100 16930 22440
H=
AP
V,
RT-
The determined Henry's constants are listed in Tables 1-111. The results on the oxygen solubility in the liquid mixtures (Table I I I ) are illustrated in Figure 1. Henry's constant runs through a minimum in the cyclohexane/benzenesystem whereas in the ethanol/toluene system a maximum is encountered. The results could be described in the form (6)
-
"
0
l
0.2
0.4
0.6
0.8
1
'ethanol
"G
Results and Dlscusslon
+ x p In H,
I
'cyclohexane
The mean values of the oxygen and carbon dioxide solubilities in water compiled in ref 1were reproduced with f l % accuracy. The reproducibility in the case of organic liquids was usually also f 1% ; at high vapor pressure of the liquid, deviations up to f2% were encountered in few cases. Each reported value is the mean of three determinations.
In H = x 1 In H,
TableIV. Me Relative Deviations of Lit rature Data d Model Predictions from Experimental Results i n Tables I and I1 mean relative deviation," % regular-solution scaled-particle gas lit data (I) theory (6) theory (7) H2 7.4 (16) 16.7 (10) 7.2 (9) He 9.5 (2) 63.8 (2) N2 4.9 (2) 10.6 (1) 27.9 (2) co 5.7 (2)b 16.4 (1) 0 2 6.3 (12) 12.0 (10) 52.7 (9) Ar 8.4 (2) 2.7 (1) 67.8 (2) COZ 32.1 (10) 52.2 (9) mean 7.0 (36) 19.2 (33) 40.2 (33) "Number of G/L systems in parentheses. bReported by Wilhelm and Battino (8).
saturated with the gas within 3-6 min. The decrease of the gas pressure at isochoric and isothermal conditions was measured with a micromanometer (MDC-FC-00 1, Furness Controls Ltd.) and recorded. A mercury manometer was used in case of very high gas solubility ((20,). The changes in the composition of the liquid mixtures during evacuation were detected by refractometry. When the oxygen solubilities in aniline and tetralin were studied, a slow pressure decrease due to autoxidation occurred, but it was negligible during the short measuring time. More experimental details are reported elsewhere (4). From the total pressure decrease AP, Henry's constant H was evaluated according to the relation P o - A P - P,
No. 2, 1989 251
CY,,@^^,
(2)
H , and H 2 are Henry's constants in the pure liquids 1 and 2, respectively, and x and x 2 are the mole fractions of the liquids
Figure 1. Henry's constants for oxygen in mixtures of cyclohexane and benzene (0)and ethanol and toluene (A)at 293.2 K (-, eq 2).
in the mixture. The expansion factor was fitted to the experimental data by nonlinear regression analysis. The optimum values of +0.41 for cyclohexane/benzene and -0.14 for ethanoi/toluene describe the measured oxygen solubiliis rather well except for low ethanol mole fractions (Figure 1). Henry's constants in the pure liquids are compared to the available literature data ( 1, 8) in Table IV. The deviations are nonsystematic and amount to an average of 7 % . (The oxygen solublliies in aniline and ethyl acetate reported by Schllipfer et al. (5) deviate by +69 % and -30 % , respectively, and are not considered.) Solubility data for COPwere available not for the temperature of 293.2 K but for 298.15 K. The mean deviation of 9.2% is therefore systematic, but it is still small as compared to the errors of the two models listed in Table IV. The strong specific interactions of COPwith the liquids are obviously not suff icientiy accounted for. For most gases the regular-solution theory (6) agrees better with the observations than the scaled-particle theory (7). However, one has to consider that the comparison is not for the same gas/liiuid systems: a major
J. Chem. Eng. Data 1989, 34, 252-257
252
T a b l e V. E f f e c t o f t h e L e n n a r d - J o n e s P a r a m e t e r s f o r t h e Gas on t h e D e s c r i p t i o n o f t h e M e a s u r e d O x y g e n S o l u b i l i t i e s by t h e S c a l e d - P a r t i c l e T h e o r y
3.460 3.467
clk. K 118.0 106.7
2.902
73.4
10'Oa. m
source
ref 9 r e f 10 optimization
mean devn. 70
x
2.2
Acknowledgment We thank Prof. W.-D. Deckwer for helpful discussions.
Glossary H Henry's constant, Pa m3 mol-' k Boltzmann constant, J K-' AP pressure change by absorption, Pa PO barometric pressure, Pa vapor pressure of liquid, Pa PL R gas constant, J K-' mol-'
expansion factor (eq 2) Lennard-Jones energy, J K-' Lennard-Jones distance, m Reetstry NO. H, 1333-74-0; He,7440-597; N, 772737-9; CO. 630-08-0;
Q1.2
E c 7
advantage of the scaled-particle theory is its applicability to polar solvents. In the case of hydrogen, the predictions of the scaled-particle theory are very close to the experimental results. I t was tried to get a better fit also for oxygen by modification of the gasspecific Lennard-Jones parameters (Table V). I f the parameters of Hirschfelder et al. (9)used by Battino and Wilhelm (7) are substituted by those suggested by Reid et al. (70),the mean error of the estimated oxygen solubilities is reduced from 52.7% to 18.8%. Fitting the two parameters to the experimental data but still using the same liquid-specific parameters, the mean error is only 2.2%. This is already close to the experimental error. I t is therefore considered to empirically modify the scaled-particle theory by fitting also the gas-specific parameters to a larger set of experimental gas solubilities.
1(2)
Greek Letters
52.7
18.8
temperature, K gas (liquid) volume, m3 mole fraction of component 1 (2) in binary mixtures
T VqLl
0, 7782-44-7; CO, 124-38-9; acetone, 67-64-1; aniline, 62-53-3; benzene, 7 1-43-2; 1-butanol, 71-36-3; n-butyl acetate, 123-86-4; carbon tetrachloride, 56-23-5; cyclohexane, 110-82-7; decalin, 9 1-17-8; 1,2dlchloroethane, 107-06-2; l,Mioxane, 123-91-1; ethanol, 64-17-5; ethyl acetate, 141-78-6; ethylbenzene, 100-41-4; methanol, 67-56-1; nitrobenzene, 9895-3; 2-propanol, 67-63-0; 1-tetradecene, 1 120-36-1; tetralin, 119-64-2; 1,2,4-trimethyIbenzene, 95-63-6; toluene, 108-88-3; m-xylene, 108-38-3.
Literature Clted Solubility Data Series; Kertes, A. S . , Ed.; Pergamon Press: Oxford, 1981-85; Vol. 1, 4, 5/6, 7, and 10. Schumpe, A.; Luhrlng, P. J. Chem. Eng. Data, submitted for publlcation. Schumpe, A.; Quicker, G.; Deckwer, W.-D. Adv. Bbchem. Eng. 1082, 2 4 , 1. Luhring, P.; Schumpe, A. Chem. Ing. Tech. 1086, 58, 976. Schgpfer, P.; Audykowski, T.; Bukowieckl, A. Schweiz. Arch. Angew. Wiss. Tech. 1040, 15, 299. Hildebrand, J. H.; Prausnitz, J. M.; Scott, R. L. Regular 8nd Rekted Solutions; Van Nostrand Reinhold: New York, 1970; Chapter 8. Battino, R.; Wilheim, E. J. Chem. Thermodyn. 1071, 3, 409. Wllhelm, E.; Battino, R. Chem. Rev. 1073, 73, 5. Hirschfekler, J. 0.; Curtis, C. F.; Bird, R. B. Molecular Theory of Gases 8nd Liquids; Wlley: New York, 1966; Chapter 13. Reid, R. C.; Prausnitz, J. M.; Sherwood, T. K. The Properties of Gases and Liquids; McGraw-Hill: New York, 1977; p 679. Received for review August 11, 1988. Accepted February 7, 1989. Financial support by the Deutsche Forschungsgemelnschaff is gratefully acknowledged.
Representation of Vapor-Liquid Equilibrium Data for Binary Refrigerant Mixtures Krister Strom, Urban Gren, * and Kjell Ljungkvist Department of Chemical Engineering Design, Chalmers University of Technology, Gijteborg, Sweden
Vapor-llquld measurements were performed on binary refrigerant mlxtures by using a Jones circulation unit wlth circulated vapor phase. Resutts from the equlllbrium measurements on the binary refrlgerants CCIzFz/C2C12F4, CHCIF2/C,CI,F4, and CHCIF2/CCI,F2 In the pressure range 3.50-14.50 bar are presented. Comparlsons were made between experimental equilbrium data and theoretlcally calculated data based on the followlng equatkms of state: Soave-Red-Kwong, de Santls, and Lee-Kesler-Pliiclter. Good agreement was obtained between the modifled Redllch-Kwong equation of state by Soave and the Lee-Kesier-Mocker equatlon of state. I n some cases and within small concentratlon intervals, the de S a n a equatlon of state was found to give divergences. This may be explained by the uncertalnty In the temperature-dependent parameters determined from pressure-volume-temperature data at temperatures near the critical.
* Author to whom correspondence should be addressed.
Introduction I t is necessary to have relevant experimental data in order to make a comparison possible between different theoretical models that describe vapor-liquid equilibrium properties. An area of increasing interest is that of binary systems of refrigerant mixtures. However, such data have been published only to a very limited extent ( 1 , 2). I n the present study isobaric measurements of vapor-liquid equilibrium data were performed for the three systems CCI,F2/C2CI2F,, CHCIF2/C2CI2F,, and CHCIF,/CCI,F,, in the pressure range 3.5-14.5 bar.
Theory The conditions for equilibrium between the vapor and the liquid phase can be expressed as f/V
=
f,L
(1)
where f,' is the fugacity of component i in the vapor phase and 4L is the fugacity of component i in the liquid phase. The fugacity of component i in a mixture is related to the pressure P and the composition z, by the fugacity coefficient according to
0021-9568/89/1734-0252$01.50/00 1989 American Chemical Society